Journal of Arrhythmia 28 (2012) 83–90

Contents lists available at SciVerse ScienceDirect

Journal of Arrhythmia

1880-42

http://d

n Corr

E-m

journal homepage: www.elsevier.com/locate/joa

Review

Catheter ablation of atrial fibrillation: Past, present, and future directions

Kentaro Yoshida, MD, PhDn, Kazutaka Aonuma, MD, PhD

Division of Cardiovascular Medicine, University of Tsukuba, 1-1-1 Tennodai, Tsukuba 305-8575, Japan

a r t i c l e i n f o

Article history:

Received 6 March 2012

Accepted 6 March 2012Available online 7 April 2012

Keywords:

Atrial fibrillation

Catheter ablation

76/$ - see front matter & 2012 Japanese Hea

x.doi.org/10.1016/j.joa.2012.03.003

esponding author. Tel.: þ81 29 853 3142; fax

ail address: kentaroyo@nifty.com (K. Yoshida

a b s t r a c t

Atrial fibrillation (AF) is the most common cardiac arrhythmia encountered in clinical practice. Because

of inadequate efficacy of pharmacological therapy, catheter ablation of AF has evolved dramatically

over the last decade. Although the success rate of ablation has improved, the ablation strategy is still

extensive, and the ablation procedure is technically challenging. In the past decade, electrophysiologists

were eager to obtain high success rates with extensive ablation. In the present decade, further

clarification of the complex mechanism of AF is required to make ablation of AF safer and much more

efficient. Because the mechanism of AF is very complex, and even somewhat mysterious, it may not be

easy to attain a better understanding of the mechanism involved or to discover better guidance for

catheter ablation. However, it is important to note that research into AF leads to better understanding

of other cardiac and non-cardiac diseases because AF develops multifactorially in association with

underlying systemic pathophysiologies.

& 2012 Japanese Heart Rhythm Society. Published by Elsevier B.V. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83

2. Patient selection according to the current guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84

3. Patient selection and reverse remodeling after ablation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84

4. History of ablation strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84

5. Do we need complete isolation of all PVs in every patient? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84

6. Relation between hemodynamic status and AF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85

7. Advances in technology to isolate PVs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85

8. Ablation of persistent AF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85

8.1. Efficacy of ablation of complex fractionated atrial electrograms (CFAEs) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85

8.2. Atrial tachycardia and efficacy of linear ablation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

8.3. Ablation endpoint . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

9. The mechanism of persistent AF from the procedural point of view. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

10. Complications. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87

11. Future direction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88

Conflict of interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88

1. Introduction

Atrial fibrillation (AF) is the most common cardiac arrhythmiaencountered in clinical practice. The number of people with AFin the United States is currently estimated at 2.4 million, and theprojected number with AF may exceed 10 million by 2050 [1,2].

rt Rhythm Society. Published by E

: þ81 29 853 3143.

).

Treatment for AF is also an important societal issue in that itrepresents a significant health care cost, currently estimated to beabout h13.5 billion annually in the European Union [4]. Increasedstroke risk in the presence of AF, reported to be an almost fivefoldexcess [5], makes AF more than a simple cardiovascular disease.

Because of inadequate efficacy of pharmacological therapy,catheter ablation of AF has evolved dramatically over the lastdecade. Although the success rate of ablation has improved toZ80% with multiple procedures, ablation strategy is still exten-sive and the ablation procedure is technically challenging,

lsevier B.V. All rights reserved.

K. Yoshida, K. Aonuma / Journal of Arrhythmia 28 (2012) 83–9084

resulting in a very high rate, around 50%, of redo procedures. Thepast decade was an era when electrophysiologists were eager toobtain high success rates with extensive ablation. In the presentdecade, we are required to further clarify the complex mechan-isms of AF to make ablation of AF safer and much more efficient.The ultimate idea is to precisely tailor ablation strategy to theparticular AF mechanism of the patient.

Fig. 1. Circumferential pulmonary vein isolation. The 3-dimensional geometry of

the left atrium and pulmonary veins was constructed with the CARTO system.

2. Patient selection according to the current guidelines

It is important to recognize that the primary justification for anAF ablation procedure at this time is the presence of symptomatic

AF, with the goal of improving patient quality of life [6]. Althoughother reasons for performing AF ablation may be justified, such asto decrease stroke risk [7], decrease heart failure risk, and improvesurvival, they have not yet been systematically evaluated as part ofa large randomized clinical trial and are therefore unproven. In theAFFIRM trial, there were no significant differences in the all-causedeaths between rhythm control and rate control using antiar-rhythmic drug therapy [8]. However, the beneficial effect on thesurvival of restoration of sinus rhythm might be offset by theadverse effects of antiarrhythmic drugs. Therefore, sinus rhythmmay be preferred over rate control if it can be achieved by catheterablation. Large prospective multicenter randomized clinical trialswill be needed to definitively determine whether sinus rhythmachieved with ablation techniques lowers morbidity and mortalityas compared with rate control alone or treatment with antiar-rhythmic therapy.

3. Patient selection and reverse remodeling after ablation

Left atrial (LA) size has been established as a prognostic marker ofcardiovascular morbidity, mortality, and stroke [9,10]. Studies haveshown that LA enlargement and function can improve, i.e., ‘‘reverseremodeling,’’ after restoration of sinus rhythm from AF with certainmedications or catheter interventions, including radiofrequency abla-tion [11–13]. A reduction in LA volume may lead to the improvementin LA function and exercise tolerance [14,15], decreased likelihood ofthrombus formation [16], and decreased susceptibility to furtheratrial arrhythmias [13,17]. Although these benefits need to be provenby large randomized prospective trials, reverse remodeling afterablation appears to justify a more aggressive clinical approach evenin less symptomatic patients. Currently, 2 studies have reportedechocardiographic predictors, including LA strain and strain rate, ofLA reverse remodeling after ablation [11,13]. It may be meaningful toselect patients who are likely to achieve reverse remodeling assuitable candidates for catheter ablation of AF.

4. History of ablation strategies

After Haı̈ssaguerre et al. reported that the ectopies frompulmonary veins (PVs) are responsible for the initiation of AF,eliminating triggers from the PVs was emphasized as a reasonableapproach to treat AF. However, this approach, directly targetingfocal triggers, was fraught with high recurrence rates due to theinfrequency with which AF initiation could be reproduciblytriggered during the catheter ablation procedure, as well as anattendant small risk of PV stenosis.

Segmental ostial ablation was the first catheter-based techni-que found to electrically isolate the PVs [18,19]. Ablation wasperformed at the ostia of the PVs, and the acute endpoint of PVisolation (PVI) could be reached in nearly every patient. However,the long-term success rates are relatively modest (60–70%).Circumferential PVI (almost equal to wide-area antral PVI), which

involves creating circumferential lesions at the PV antra aroundthe ipsilateral PVs, improved outcomes in patients with bothparoxysmal and persistent AF (Fig. 1) [20]. The superiority ofcircumferential PVI over segmental PVI can be explained by thefollowing: circumferential PVI may extinguish the triggers anddrivers located in less common trigger sites other than the PVs,including the antral region of the PVs, the vein and ligament ofMarshall, and the posterior LA wall. Circumferential PVI mayimpact not only triggers but also the arrhythmogenic substratestabilizing the maintenance of AF [21]. The reduction of atrialmuscle mass may make coexistence of multiple reentries impos-sible (debulking effect) [22]. Moreover, circumferential PVI mayinterrupt sympathetic and parasympathetic innervation from theautonomic ganglia, which have been identified as potentialtriggers for AF [23]. In a series of 349 consecutive patients under-going circumferential PVI at the University of Michigan, AF waseliminated in 87% of patients with paroxysmal AF and 75% ofpatients with persistent AF [24].

5. Do we need complete isolation of all PVs in every patient?

Circumferential PVI results in satisfactory outcomes for ablationof paroxysmal AF, and electrical isolation of all PVs is currently astandard approach to the treatment of AF, as recommended by theexpert consensus statement [6]. Complete isolation also seems tobe important for preventing recurrent atrial tachycardia duringfollow-up [25–27]. However, it is not clear whether all patientswith AF need to undergo isolation of all PVs. A number of studieshave reported favorable outcomes with an ablation strategy thatdoes not include PVI. Lemola et al. reported that a successfuloutcome after LA ablation was found to be independent of thenumber of PVs that were electrically isolated. Therefore, completeisolation of the PVs was not necessary for a successful outcome[21]. Oral et al. reported that a tailored ablation strategy thattargeted driver tachycardias and complex electrograms in theselected PVs resulted in freedom from recurrent AF in �80% ofpatients with paroxysmal AF [28]. Pokushalov et al. reported thatselective ganglionated plexi ablation directed by an anatomicapproach resulted in successful outcomes in �80% of patients withparoxysmal AF [29]. In the Pratola et al. study [29], patients withpersistent AF who underwent PVI and did not have AF recurrenceunderwent repeated electrophysiological studies. Notably, PVIpersisted in only �40% of the previously isolated PVs. This studysupported the idea that atrial substrate also plays an important role

K. Yoshida, K. Aonuma / Journal of Arrhythmia 28 (2012) 83–90 85

in the persistence of AF, and the presence of PV reconnection doesnot necessarily mean recurrence of AF [30].

In contrast, there is consensus that in the majority of patients,AF recurs in association with recovered PV conduction. In a longfollow-up study (median follow-up period of 4.8 years), recoveredPV conduction was observed in 62 of 66 patients (94%) at the 2ndablation procedure [31]. The discrepancy in the impact of PVI onthe maintenance of sinus rhythm may be explained by theinteraction of a trigger and a substrate. The presence of AFsubstrate in addition to PV reconnection makes the atria moresusceptible to recurrent AF than does PV reconnection alone.However, little data directly supports this notion because theactual prevalence of PV reconnections in patients free fromrecurrence of AF cannot be evaluated in routine clinical practice.

These studies have indicated that aiming for isolation of all PVsmight be an excessive treatment for some patients. However, thecritical issue is that during the procedure we never completelyknow which PV is the trigger that initiates AF (the so-calledarrhythmogenic PV) and which PV is not arrhythmogenic. If wecould determine this, the efficiency of AF ablation would improvedramatically. Advances in technology to make achievement ofisolation of all PVs (so-called prophylactic PVI) safer and easier,especially in paroxysmal AF, may resolve the issue that thearrhythmogenic PV critical to the initiation and perpetuation ofAF cannot be definitively determined during the procedure.

6. Relation between hemodynamic status and AF

AF and heart failure create a vicious circle: heart failurepromotes AF, and AF aggravates heart failure. In patients withsymptomatic heart failure, the prevalence of AF ranges from 10 to30% [32]. The importance of atrial stretch associated with anincrease in atrial pressure in the maintenance of AF has beenreported in animal models [33] and in patients with AF [34]. Inthe human study, patients with persistent AF had significantlyhigher LA pressures than did patients with paroxysmal AF. Theatrial activation rate is known to be higher in patients withpersistent AF than in patients with paroxysmal AF. Higher LApressure may result in a greater degree of stretch-related elec-trical remodeling and an increase in atrial activation rate, makingspontaneous termination of AF less likely.

A potential new risk factor for AF, stiff LA syndrome, wasrecently proposed. The syndrome itself was originally reportedin the late 1980s [35]. Its principal feature is that right heartfailure is disproportionate to left heart failure because ofreduced LA compliance or LA diastolic dysfunction. Recentstudies have indicated that one of the causes of the syndromewas LA ablation for AF [36,37]. Machino-Ohtsuka et al. reportedthat pre-existing LA stiffness was related to AF recurrence afterablation [38]. Using magnetocardiography analysis, Sato et al.showed that right atrial overload possibly due to decreased LAcompliance after LA ablation was associated with AF recurrenceafter ablation [39].

Although a well experienced group reported a very highsuccess rate of 80% after catheter ablation even in patients withcongestive heart failure [40], the evidence described above suggeststhat catheter ablation alone is not sufficient to treat AF in patientswith hemodynamic deterioration. This indicates that pharmacolo-gical therapies for heart failure, underlying cardiac pathophysiol-ogy, and hypertension, such as angiotensin-converting enzymeinhibitors, beta-blockers, statins, and diuretics, are as important ascatheter ablation to decrease susceptibility to AF and to preventrecurrence of AF after ablation. Obesity and sleep apnea also affecthemodynamic status and are associated with the prognosis of AF.Continuous positive air pressure therapy (CPAP) may be effective

for maintenance of sinus rhythm after ablation [41–43], but thisneeds to be tested by randomized prospective trials.

7. Advances in technology to isolate PVs

Standard 4-mm-tip ablation catheters were initially used forLA ablation. Because of the unstable and limited energy deliveryof 4-mm-tip catheters due to the temperature-limited setting,8-mm-tip catheters were introduced. However, one of problemsin the use of 8-mm-tip catheters was the ambiguity of localelectrograms due to the inclusion of far-field electrograms. Since2000, irrigated-tip catheters have been used in European coun-tries, and they offer several advantages, including delivery of thedesired power independent of local blood flow [44]. The useful-ness of this technology has been supported in animal models [45]and in patients undergoing ablation of AF [46,47].

Efforts continue to improve the efficacy and safety of LAablation. Traditional catheter ablation is performed as a single-tip, point-by-point ablation process. This technique requires ahigh degree of operator skill, and procedures are lengthy, oftenlasting more than 4 h. Creating reliable continuous transmurallesions with a single-point catheter is difficult. During the pastdecade, a number of alternative ablation systems have appeared,including cryoablation (with a conventional-tip catheter or acircular catheter or balloon device), ultrasound ablation, laserablation, and an over-the-wire multi-electrode catheter deliver-ing duty-cycled bipolar and unipolar RF energy. A radiofrequencyablation catheter capable of real-time tissue–tip contact forcemeasurements has recently been developed and has garneredparticular attention. The contact force between the catheter tipand the tissue may affect the clinical outcome of RF ablation forthe treatment of cardiac arrhythmias [48,49]. Insufficient contactforce may result in an ineffective lesion, whereas excessivecontact force may result in complications. Preliminary studieshave indicated that a contact-force sensing catheter is useful forsafe catheter manipulation and reduction of fluoroscopy time. Inthe future, it may also increase the effectiveness of ablations byallowing better control of the lesion size.

Although some of the modalities described are not currentlyavailable in Japan, we should keep in mind that advances intechnology are as important as is understanding of the AFmechanism in clinical practice.

8. Ablation of persistent AF

Because triggers/drivers that originate from the PVs and otherthoracic veins appear to be the primary mechanism of parox-ysmal AF, ablation strategies that target only thoracic veinarrhythmogenicity have been effective in the majority of patientswith paroxysmal AF. Additional linear ablation in combinationwith circumferential PVI resulted in an increased incidence of LAatrial tachycardia compared with circumferential PVI alone inpatients with paroxysmal AF [50]. However, elimination of PVarrhythmogenicity alone has been insufficient to eliminate per-sistent AF. Although catheter ablation has also evolved into aneffective treatment strategy in patients with persistent AF[46,47,51,52], the mechanism of persistent AF is still unclearand ablation efficiency remains low.

8.1. Efficacy of ablation of complex fractionated atrial electrograms

(CFAEs)

Nademanee et al. performed ablation of AF that targets CFAEs,which are defined as electrograms with a cycle length of more

K. Yoshida, K. Aonuma / Journal of Arrhythmia 28 (2012) 83–9086

than 120 ms, or shorter than that in the coronary sinus, or thosethat were fractionated or displayed continuous electrical activity[53]. Ablation of CFAEs resulted in termination of AF in �95% ofpatients and an excellent clinical outcome, in which 90% ofpatients were reported to be free from recurrent arrhythmias.This ablation strategy is currently applied in combination withPVI because the clinical efficacy of ablation of CFAEs alone hasbeen only modest, except in one study [53]. CFAE ablation inconjunction with PV antral isolation has a higher likelihood ofmaintaining sinus rhythm compared with PV antral isolationalone in patients with persistent AF [51]. Another randomizedstudy showed that ablation of CFAEs after PV antral isolation didnot have any incremental value over PV antral isolation alone[54]. The major problem concerning CFAE ablation is that thedefinition of CFAEs is arbitrary and subjective. The coexistence ofthe two fundamentally different types of electrograms thatsuggest different etiologies in one definition may not be appro-priate for targeting a particular mechanism of AF. Although theauthors of some studies tried to objectively detect CFAEs withcustom software and demonstrate differences in the distributionof CFAEs in the LA between paroxysmal and persistent AF [55,56],the impact of the objective definition on clinical outcomes isunclear. The very high prevalence of CFAEs in the left atriumsuggests that CEAEs alone are a nonspecific marker of appropriatetarget sites for ablation of AF [57]. Therefore, the value of CFAEsablation remains controversial.

8.2. Atrial tachycardia and efficacy of linear ablation

Atrial tachycardia is common after catheter ablation of AF.Among a large number of patients undergoing circumferentialPVI, the prevalence of atrial tachycardia at follow-up wasreported to be 24% [24]. Forty-seven of 50 (94%) consecutivepatients with mitral isthmus-dependent flutter following orduring AF ablation had persistent AF as their primary arrhythmia[58]. Persistent AF is much more likely to convert to atrialtachycardia rather than to sinus rhythm during ablation. In 100patients with persistent AF, atrial tachycardia comprised one-third of the recurrent arrhythmias after PV antral isolation incombination with CFAE ablation, and, notably, if patients had acritical decrease in the dominant frequency (DF) of AF of greaterthan 11% or had termination of AF during the procedure, the ratio ofatrial tachycardia to total recurrent arrhythmia increased to 70% [3].

The majority of atrial tachycardias that occurred after circum-ferential PVI for AF were caused by a re-entrant mechanism.Mitral isthmus, roof, and septum accounted for 75% of theablation target sites for macro-reentrant atrial tachycardias fromthe left atrium. The critical isthmus in 115 of 120 (96%) LA re-entrant atrial tachycardias traversed a prior ablation line, indicat-ing that they were gap related [25]. There is, however, a hypoth-esis that the atrial tachycardia that occurs during catheterablation of AF is a driver of AF that manifests after eliminationof fibrillatory conduction [59]. Because both reports suggested astrong correlation between persistent AF and atrial tachycardia,prophylactic linear ablation accompanied by PVI may improveablation outcome irrespective of whether AT is a cause or aneffect. Retrospective evaluation found that linear ablations formacro-reentrant atrial tachycardia were required in the majorityof patients with persistent AF even if persistent AF was termi-nated without linear ablation in the index procedure [60]. How-ever, complete bidirectional block of the mitral isthmus usuallyrequires aggressive ablation with a combined endocardial andepicardial approach. Interposition of the circumflex arterybetween the mitral isthmus and the coronary sinus is associatedwith a lower probability of achieving complete mitral isthmusblock [61,62]. Particularly in such cases, this may result in a

higher risk of complications such as cardiac tamponade anddamage to the circumflex artery [63]. Assessment of the incre-mental value of linear ablations after PVI requires a randomizedprospective study.

8.3. Ablation endpoint

Termination of AF during ablation is predictive of freedomfrom recurrent AF [46,47,64,65]. However, termination of persis-tent AF usually requires extensive ablation beyond the PVs,including ablation of CFAEs and multiple linear lesions. Extensiveablation is associated with long procedure time, radiation expo-sure, proarrhythmia, risk of collateral damage, compromise of LAtransport function, and stiff LA syndrome. A multicenter prospec-tive study showed the conflicting result that AF cycle length atbaseline and termination of AF during ablation were not pre-dictive of long-term sinus rhythm maintenance [66]. Anotherstudy showed that patient age and duration of radiofrequencyenergy were independent predictors of the outcomes after multi-ple procedures, whether or not the AF terminated during theprocedure [67]. In a retrospective analysis of consecutive patientswith persistent AF at the University of Michigan, a decrease in thedominant frequency of AF by 11% in response to PV antral isolationand ablation of CFAEs was associated with a probability of main-taining sinus rhythm that was similar to that when radiofrequencyablation terminated AF (Fig. 2) [3]. Appropriate patient selectionand an endpoint tailored to each patient are essential for improve-ment of efficiency of catheter ablation of persistent AF.

9. The mechanism of persistent AF from the procedural pointof view

The main explanation for the current disappointing ability tocontrol AF is an incomplete understanding of the mechanismunderlying its maintenance, despite many years of research andspeculation. Over the past 50 years, the multiple wavelet hypoth-esis has been the dominant mechanistic model of AF. Thishypothesis, first postulated by Moe et al., states that AF is theresult of randomly propagating multiple electrical wavelets thatinteract in very complex ways, with local excitation limited by theheterogeneous distribution of refractory periods throughout theatria [68]. According to this model, the number of wavelets at anypoint in time depends on the atrial conduction velocity, refractoryperiod, and mass. Perpetuation of AF is favored by slowedconduction, shortened refractory periods, and increased atrialmass. The shorter the wavelength is, the higher the number ofwavelets there are. The presence of more wavelets makes perpe-tuation of AF more stable.

The so-called driver hypothesis has been posited in the recentyears. Optical mapping studies in isolated sheep hearts havesuggested that at least some cases of AF can be maintained byhigh-frequency reentrant sources (rotors), usually located in theposterior LA, which result in spatially distributed frequencygradients [69,70]. High-frequency rotors maintain AF throughfibrillatory conduction to the remainder of the atria. From theclinical point of view, in patients with paroxysmal AF, thishypothesis is supported by the fact that high-frequency activitywithin the PV continues even after the restoration of sinus rhythmin the atria [71]. Presence of a left-to-right atrial frequency gradientin patients with paroxysmal AF also supports this hypothesis [72]. Inpersistent AF, however, electrophysiologists experience someresponses of AF dynamics to ablation that are not consistent withthe findings estimated from the driver hypothesis. For example,there is no left-to-right atrial frequency gradient in patients withpersistent AF [72]. It is rare to find sites of DFmax in the PV region,

Fig. 2. Spectral analysis of atrial fibrillation. Electrograms recorded from the coronary sinus and lead V1 were processed, and the dominant frequency (DF) was determined

by fast Fourier transformation. Shown are the 12-lead ECG (Panel A) and the periodograms (Panel B) at baseline and after radiofrequency catheter ablation. The DF in the

coronary sinus (CS) decreased from 7.03 Hz to 6.21 Hz after ablation in this patient [58].

K. Yoshida, K. Aonuma / Journal of Arrhythmia 28 (2012) 83–90 87

and isolated PV tachycardia after restoration of sinus rhythm in theatria has never been observed. A cumulative effect is frequentlyobserved in ablation of persistent AF. Haı̈ssaguerre et al. reportedthat although stepwise ablation was performed in a randomizedorder (PVI, atrial ablation, and coronary sinus/superior vena cavaablation), the number of patients with termination of AF increasedas these procedural steps were completed [47]. It is a commonfinding that sites where radiofrequency application terminated AFhad been previously ablated [73]. Taken together, it is difficult forthe driver hypothesis alone to explain the findings observed during

the procedure. One of the possibilities that can account for thesefindings might be the presence of multiple AF drivers, but this hasnot been proven in humans.

10. Complications

Because the major purpose of catheter ablation of AF is toimprove patients’ quality of life, and its efficacy in improvingsurvival has not been proven, the probability and severity of

K. Yoshida, K. Aonuma / Journal of Arrhythmia 28 (2012) 83–9088

complications related to AF ablation must be recognized by allelectrophysiologists, and patients should only undergo AF abla-tion after carefully weighing the risks and benefits of theprocedure. Because AF ablation is one of the most complexinterventional procedures performed, the risk associated withthis procedure is estimated to be higher than that of theprocedures to treat other arrhythmias. Although prospectivestudies surveying the mortality rates associated with AF ablationprocedures are lacking, a worldwide retrospective survey ofprevalence and causes of fatal outcomes in AF ablation reportedthat lethal adverse events occurred in 0.1% of patients undergoingAF ablation [74]. Major causes of fatal outcomes were tamponade,stroke, and atrioesophageal fistula. More recently, predictors ofcomplications and 30-day readmissions were identified from thedata from the California State Inpatient Database [75]. Fivepercent of patients experienced periprocedural complications,and almost 10% were rehospitalized within 30 days. One patientdied during the index admission, and 9 patients died during the30-day rehospitalization, resulting in a mortality rate of 0.24%.These rates are quite high when one considers that AF itself is nota life-threatening arrhythmia.

Because a delay in the diagnosis of tamponade is fatal,continuous monitoring of systemic arterial pressure during andfollowing AF ablation is mandatory. Although the majority ofepisodes of tamponade can be managed by immediate percuta-neous drainage, surgical drainage and repair are sometimesneeded. Thus, AF ablation should only be performed in hospitalsequipped to provide emergency surgical support when required.

Oral et al. reported that AF ablation is associated with earlypostprocedure thromboembolism, regardless of both the post-procedure rhythm and whether the patient has risk factors forstroke. The most likely cause was thought to be char and/orthrombus formation at sites of LA endocardial ablation, and theprobability was 1.0% [76]. Thus, heparin anticoagulation withclose attention to maintaining a therapeutic dose (activatedclotting time (ACT) of at least 300–350 s) during the procedureis important. Data regarding the risk of thromboembolism withand without warfarin after AF ablation is limited. Becausesymptomatic or asymptomatic AF may recur during long-termfollow-up after an AF ablation procedure [31], discontinuation ofwarfarin therapy post-ablation generally is not recommended inpatients who have a CHADS2 score Z2 [6]. Although the use ofdabigatran has increased in clinical practice, an observationalstudy with a matched-control design reported that periproceduraldabigatran use significantly increased the risk of bleeding orthromboembolic complications compared with uninterruptedwarfarin therapy in patients undergoing AF ablation [77]. Largerandomized controlled studies are required to confirm this result.

Development of an atrial–esophageal fistula is one of the mostdreaded complications of AF ablation [78]. A relatively large-scalenonrandomized study revealed that the anatomical risk factor of asmall LA-to-esophageal distance was the most important factor inesophageal ulceration when using an irrigation-tip catheter at anenergy setting of �25 W at the posterior left atrium. Most of thepatients who developed an atrial–esophageal fistula in this studydied (7/9 patients, 78%) [74], and it is vital to avoid thiscomplication. The most common practice is to decrease powerdelivery, decrease tissue contact pressure, and move the ablationcatheter every 10–20 s when in close proximity to the esophagus.A number of other approaches are also used to avoid thedevelopment of an atrial–esophageal fistula [79–84], includingtemperature monitoring of the esophagus, use of pain as an assayfor potential esophageal injury, the use of capsule endoscopy afterAF ablation, and mechanical displacement of the esophagusduring the AF procedure. Yamasaki et al. showed that low bodymass index is a predictor for esophageal injury even at low energy

settings of radiofrequency delivery [85]. Because Asian people aregenerally thinner than people in other regions of the world, therisk of esophageal injury must always be considered in eachpatient undergoing ablation of AF.

11. Future direction

The mechanisms of AF are very complex and even somewhatmysterious. While it may not be easy to attain a greater under-standing of the mechanism of AF or to discover clearer guidanceon catheter ablation, research on AF nonetheless leads to a betterunderstanding of other cardiac and non-cardiac diseases becauseAF develops multifactorially in association with various under-lying systemic pathophysiologies. To understand AF is to under-stand the whole body. It is our goal and hope as a next step topromote catheter ablation as a first-line therapy in more patientswith AF based on definitive evidence that the procedure reducesmortality and morbidity.

Conflict of interest

All authors have no conflict of interest to declare.

References

[1] Miyasaka Y, Barnes ME, Gersh BJ, et al. Secular trends in incidence of atrialfibrillation in Olmsted County, Minnesota, 1980 to 2000, and implications onthe projections for future prevalence. Circulation 2006;114:119–25.

[2] Naccarelli GV, Varker H, Lin J, et al. Increasing prevalence of atrial fibrillationand flutter in the United States. Am J Cardiol 2009;104:1534–9.

[3] Yoshida K, Chugh A, Good E, et al. A critical decrease in dominant frequencyand clinical outcome after catheter ablation of persistent atrial fibrillation.Heart Rhythm;7:295–302.

[4] Le Heuzey JY, Paziaud O, Piot O, et al. Cost of care distribution in atrialfibrillation patients: the COCAF study. Am Heart J 2004;147:121–6.

[5] Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent riskfactor for stroke: the Framingham Study. Stroke 1991;22:983–8.

[6] Calkins H, Brugada J, Packer DL, et al. HRS/EHRA/ECAS expert ConsensusStatement on catheter and surgical ablation of atrial fibrillation: recommen-dations for personnel, policy, procedures and follow-up. A report of the HeartRhythm Society (HRS) Task Force on catheter and surgical ablation of atrialfibrillation. Heart Rhythm 2007;4:816–61.

[7] Hunter RJ, McCready J, Diab I, et al. Maintenance of sinus rhythm with anablation strategy in patients with atrial fibrillation is associated with a lowerrisk of stroke and death. Heart 2011;98:48–53.

[8] Wyse DG, Waldo AL, DiMarco JP, et al. A comparison of rate control andrhythm control in patients with atrial fibrillation. N Engl J Med 2002;347:1825–1833.

[9] Benjamin EJ, D’Agostino RB, Belanger AJ, et al. Left atrial size and the risk ofstroke and death. The Framingham Heart Study. Circulation 1995;92:835–41.

[10] Leung DY, Boyd A, Ng AA, et al. Echocardiographic evaluation of left atrialsize and function: current understanding, pathophysiologic correlates, andprognostic implications. Am Heart J 2008;156:1056–64.

[11] Kuppahally SS, Akoum N, Badger TJ, et al. Echocardiographic left atrialreverse remodeling after catheter ablation of atrial fibrillation is predicted bypreablation delayed enhancement of left atrium by magnetic resonanceimaging. Am Heart J 2010;160:877–84.

[12] Tops LF, Bax JJ, Zeppenfeld K, et al. Effect of radiofrequency catheter ablationfor atrial fibrillation on left atrial cavity size. Am J Cardiol 2006;97:1220–2.

[13] Tops LF, Delgado V, Bertini M, et al. Left atrial strain predicts reverseremodeling after catheter ablation for atrial fibrillation. J Am Coll Cardiol2011;57:324–31.

[14] Thomas L, Boyd A, Thomas SP, et al. Atrial structural remodelling andrestoration of atrial contraction after linear ablation for atrial fibrillation. EurHeart J 2003;24:1942–51.

[15] Sanders P, Morton JB, Morgan JG, et al. Reversal of atrial mechanical stunningafter cardioversion of atrial arrhythmias: implications for the mechanisms oftachycardia-mediated atrial cardiomyopathy. Circulation 2002;106:1806–13.

[16] Buber J, Luria D, Sternik L, et al. Left atrial contractile function following asuccessful modified Maze procedure at surgery and the risk for subsequentthromboembolic stroke. J Am Coll Cardiol 2011;58:1614–21.

[17] Casaclang-Verzosa G, Gersh BJ, Tsang TS. Structural and functional remodel-ing of the left atrium: clinical and therapeutic implications for atrialfibrillation. J Am Coll Cardiol 2008;51:1–11.

K. Yoshida, K. Aonuma / Journal of Arrhythmia 28 (2012) 83–90 89

[18] Oral H, Knight BP, Ozaydin M, et al. Segmental ostial ablation to isolate thepulmonary veins during atrial fibrillation: feasibility and mechanisticinsights. Circulation 2002;106:1256–62.

[19] Pappone C, Rosanio S, Oreto G, et al. Circumferential radiofrequency ablationof pulmonary vein ostia: a new anatomic approach for curing atrial fibrilla-tion. Circulation 2000;102:2619–28.

[20] Oral H, Scharf C, Chugh A, et al. Catheter ablation for paroxysmal atrialfibrillation: segmental pulmonary vein ostial ablation versus left atrialablation. Circulation 2003;108:2355–60.

[21] Lemola K, Oral H, Chugh A, et al. Pulmonary vein isolation as an end point forleft atrial circumferential ablation of atrial fibrillation. J Am Coll Cardiol2005;46:1060–6.

[22] Byrd GD, Prasad SM, Ripplinger CM, et al. Importance of geometry andrefractory period in sustaining atrial fibrillation: testing the critical masshypothesis. Circulation 2005;112:l7–13.

[23] Lemola K, Chartier D, Yeh YH, et al. Pulmonary vein region ablation inexperimental vagal atrial fibrillation: role of pulmonary veins versus auto-nomic ganglia. Circulation 2008;117:470–7.

[24] Chugh A, Oral H, Lemola K, et al. Prevalence, mechanisms, and clinicalsignificance of macroreentrant atrial tachycardia during and following leftatrial ablation for atrial fibrillation. Heart Rhythm 2005;2:464–71.

[25] Chae S, Oral H, Good E, et al. Atrial tachycardia after circumferential pulmonaryvein ablation of atrial fibrillation: mechanistic insights, results of catheter ablation,and risk factors for recurrence. J Am Coll Cardiol 2007;50:1781–7.

[26] Cummings JE, Schweikert R, Saliba W, et al. Left atrial flutter followingpulmonary vein antrum isolation with radiofrequency energy: linear lesionsor repeat isolation. J Cardiovasc Electrophysiol 2005;16:293–7.

[27] Ouyang F, Antz M, Ernst S, et al. Recovered pulmonary vein conduction as adominant factor for recurrent atrial tachyarrhythmias after complete circularisolation of the pulmonary veins: lessons from double Lasso technique.Circulation 2005;111:127–35.

[28] Oral H, Chugh A, Good E, et al. A tailored approach to catheter ablation ofparoxysmal atrial fibrillation. Circulation 2006;113:1824–31.

[29] Pokushalov E, Romanov A, Shugayev P, et al. Selective ganglionated plexiablation for paroxysmal atrial fibrillation. Heart Rhythm 2009;6:1257–64.

[30] Pratola C, Baldo E, Notarstefano P, et al. Radiofrequency ablation of atrialfibrillation: is the persistence of all intraprocedural targets necessary forlong-term maintenance of sinus rhythm? Circulation 2008;117:136–43.

[31] Ouyang F, Tilz R, Chun J, et al. Long-term results of catheter ablation inparoxysmal atrial fibrillation: lessons from a 5-year follow-up. Circulation2010;122:2368–77.

[32] Dries DL, Exner DV, Gersh BJ, et al. Atrial fibrillation is associated with anincreased risk for mortality and heart failure progression in patients withasymptomatic and symptomatic left ventricular systolic dysfunction: aretrospective analysis of the SOLVD trials. Studies of Left VentricularDysfunction. J Am Coll Cardiol 1998;32:695–703.

[33] Kalifa J, Jalife J, Zaitsev AV, et al. Intra-atrial pressure increases rate andorganization of waves emanating from the superior pulmonary veins duringatrial fibrillation. Circulation 2003;108:668–71.

[34] Yoshida K, Ulfarsson M, Oral H, et al. Left atrial pressure and dominantfrequency of atrial fibrillation in humans. Heart Rhythm 2010;8:181–7.

[35] Pilote L, Huttner I, Marpole D, et al. Stiff left atrial syndrome. Can J Cardiol1988;4:255–7.

[36] Gibson DN, Di Biase L, Mohanty P, et al. Stiff left atrial syndrome aftercatheter ablation for atrial fibrillation: clinical characterization, prevalence,and predictors. Heart Rhythm 2011;8:1364–71.

[37] Shoemaker MB, Hemnes AR, Robbins IM, et al. Left atrial hypertensionafter repeated catheter ablations for atrial fibrillation. J Am Coll Cardiol2011;57:1918–9.

[38] Machino-Ohtsuka T, Seo Y, Tada H, et al. Left atrial stiffness relates to leftventricular diastolic dysfunction and recurrence after pulmonary vein isola-tion for atrial fibrillation. J Cardiovasc Electrophysiol 2011;22:999–1006.

[39] Sato Y, Yoshida K, Ogata K, et al. An increase in right atrial magnetic strengthis a novel predictor of recurrence of atrial fibrillation after radiofrequencycatheter ablation. Circulation Journal, in press.

[40] Hsu LF, Jais P, Sanders P, et al. Catheter ablation for atrial fibrillation incongestive heart failure. N Engl J Med 2004;351:2373–83.

[41] Jongnarangsin K, Chugh A, Good E, et al. Body mass index, obstructive sleepapnea, and outcomes of catheter ablation of atrial fibrillation. J CardiovascElectrophysiol 2008;19:668–72.

[42] Tsang TS, Barnes ME, Miyasaka Y, et al. Obesity as a risk factor for the progressionof paroxysmal to permanent atrial fibrillation: a longitudinal cohort study of21 years. Eur Heart J 2008;29:2227–33.

[43] Patel D, Mohanty P, Di Biase L, et al. Safety and efficacy of pulmonary vein antralisolation in patients with obstructive sleep apnea: the impact of continuouspositive airway pressure. Circ Arrhythmia Electrophysiol 2010;3:445–51.

[44] Macle L, Jais P, Weerasooriya R, et al. Irrigated-tip catheter ablation ofpulmonary veins for treatment of atrial fibrillation. J Cardiovasc Electro-physiol 2002;13:1067–73.

[45] Nakagawa H, Yamanashi WS, Pitha JV, et al. Comparison of in vivo tissuetemperature profile and lesion geometry for radiofrequency ablation with asaline-irrigated electrode versus temperature control in a canine thighmuscle preparation. Circulation 1995;91:2264–73.

[46] Haissaguerre M, Hocini M, Sanders P, et al. Catheter ablation of long-lastingpersistent atrial fibrillation: clinical outcome and mechanisms of subsequentarrhythmias. J Cardiovasc Electrophysiol 2005;16:1138–47.

[47] Haissaguerre M, Sanders P, Hocini M, et al. Catheter ablation of long-lastingpersistent atrial fibrillation: critical structures for termination. J CardiovascElectrophysiol 2005;16:1125–37.

[48] Kuck KH, Reddy VY, Schmidt B, et al. A novel radiofrequency ablation catheterusing contact force sensing: Toccata study. Heart Rhythm 2012;9:18–23.

[49] Kerst G, Weig HJ, Weretka S, et al. Contact force-controlled zero-fluoroscopycatheter ablation of right-sided and left atrial arrhythmia substrates. HeartRhythm 2012;2012:2.

[50] Sawhney N, Anousheh R, Chen W, et al. Circumferential pulmonary veinablation with additional linear ablation results in an increased incidence ofleft atrial flutter compared with segmental pulmonary vein isolation as aninitial approach to ablation of paroxysmal atrial fibrillation. Circ ArrhythmiaElectrophysiol 2010;3:243–8.

[51] Elayi CS, Verma A, Di Biase L, et al. Ablation for longstanding permanentatrial fibrillation: results from a randomized study comparing three differentstrategies. Heart Rhythm 2008;5:1658–64.

[52] Oral H, Pappone C, Chugh A, et al. Circumferential pulmonary-vein ablationfor chronic atrial fibrillation. N Engl J Med 2006;354:934–41.

[53] Nademanee K, McKenzie J, Kosar E, et al. A new approach for catheterablation of atrial fibrillation: mapping of the electrophysiologic substrate.J Am Coll Cardiol 2004;43:2044–53.

[54] Oral H, Chugh A, Yoshida K, et al. A randomized assessment of the incrementalrole of ablation of complex fractionated atrial electrograms after antralpulmonary vein isolation for long-lasting persistent atrial fibrillation. J Am CollCardiol 2009;53:782–9.

[55] Scherr D, Dalal D, Cheema A, et al. Automated detection and characterizationof complex fractionated atrial electrograms in human left atrium duringatrial fibrillation. Heart Rhythm 2007;4:1013–20.

[56] Wu J, Estner H, Luik A, et al. Automatic 3D mapping of complex fractionatedatrial electrograms (CFAE) in patients with paroxysmal and persistent atrialfibrillation. J Cardiovasc Electrophysiol 2008;19:897–903.

[57] Tada H, Yoshida K, Chugh A, et al. Prevalence and characteristics of continuouselectrical activity in patients with paroxysmal and persistent atrial fibrillation.J Cardiovasc Electrophysiol 2008;19:606–12.

[58] Matsuo S, Wright M, Knecht S, et al. Peri-mitral atrial flutter in patients withatrial fibrillation ablation. Heart Rhythm 2010;7:2–8.

[59] Yoshida K, Chugh A, Ulfarsson M, et al. Relationship between the spectralcharacteristics of atrial fibrillation and atrial tachycardias that occur aftercatheter ablation of atrial fibrillation. Heart Rhythm 2009;6:11–7.

[60] Knecht S, Hocini M, Wright M, et al. Left atrial linear lesions are required forsuccessful treatment of persistent atrial fibrillation. Eur Heart J 2008;29:2359–66.

[61] Kurotobi T, Shimada Y, Kino N, et al. Local coronary flow is associated withan unsuccessful complete block line at the mitral isthmus in patients withatrial fibrillation. Circ Arrhythmia Electrophysiol 2011;4:838–43.

[62] Yokokawa M, Sundaram B, Garg A, et al. Impact of mitral isthmus anatomyon the likelihood of achieving linear block in patients undergoing catheterablation of persistent atrial fibrillation. Heart Rhythm 2011;8:1404–10.

[63] Wong KC, Lim C, Sadarmin PP, et al. High incidence of acute sub-clinicalcircumflex artery ‘injury’ following mitral isthmus ablation. Eur Heart J2011;32:1881–90.

[64] Haissaguerre M, Sanders P, Hocini M, et al. Changes in atrial fibrillationcycle length and inducibility during catheter ablation and their relation tooutcome. Circulation 2004;109:3007–13.

[65] Oral H, Chugh A, Good E, et al. Radiofrequency catheter ablation of chronic atrialfibrillation guided by complex electrograms. Circulation 2007;115:2606–12.

[66] Elayi CS, Di Biase L, Barrett C, et al. Atrial fibrillation termination as aprocedural endpoint during ablation in long-standing persistent atrialfibrillation. Heart Rhythm 2010;7:1216–23.

[67] Yoshida K, Rabbani AB, Oral H, et al. Left atrial volume and dominantfrequency of atrial fibrillation in patients undergoing catheter ablation ofpersistent atrial fibrillation. J Interventional Card Electrophysiol 2011;32:155–61.

[68] Moe GK, Abildskov JA. Atrial fibrillation as a self-sustaining arrhythmiaindependent of focal discharge. Am Heart J 1959;58:59–70.

[69] Jalife J, Berenfeld O, Mansour M. Mother rotors and fibrillatory conduction: amechanism of atrial fibrillation. Cardiovasc Res 2002;54:204–16.

[70] Mansour M, Mandapati R, Berenfeld O, et al. Left-to-right gradient of atrialfrequencies during acute atrial fibrillation in the isolated sheep heart.Circulation 2001;103:2631–6.

[71] Huang H, Wang X, Chun J, et al. A single pulmonary vein as electrophysio-logical substrate of paroxysmal atrial fibrillation. J Cardiovasc Electrophysiol2006;17:1193–201.

[72] Lazar S, Dixit S, Marchlinski FE, et al. Presence of left-to-right atrialfrequency gradient in paroxysmal but not persistent atrial fibrillation inhumans. Circulation 2004;110:3181–6.

[73] O’Neill MD, Jais P, Takahashi Y, et al. The stepwise ablation approach forchronic atrial fibrillation—evidence for a cumulative effect. J InterventionalCard Electrophysiol 2006;16:153–67.

[74] Cappato R, Calkins H, Chen SA, et al. Prevalence and causes of fatal outcomein catheter ablation of atrial fibrillation. J Am Coll Cardiol 2009;53:1798–803.

[75] Shah RU, Freeman JV, Shilane D, et al. Procedural complications, rehospita-lizations, and repeat procedures after catheter ablation for atrial fibrillation.J Am Coll Cardiol 2012;59:143–9.

[76] Oral H, Chugh A, Ozaydin M, et al. Risk of thromboembolic events afterpercutaneous left atrial radiofrequency ablation of atrial fibrillation. Circula-tion 2006;114:759–65.

K. Yoshida, K. Aonuma / Journal of Arrhythmia 28 (2012) 83–9090

[77] Lakkireddy D, Reddy YM, Di Biase L, et al. Feasibility and safety of dabigatranversus warfarin for periprocedural anticoagulation in patients undergoingradiofrequency ablation for atrial fibrillation results from a multicenterprospective registry. J Am Coll Cardiol, (Epub ahead of print).

[78] Pappone C, Oral H, Santinelli V, et al. Atrio–esophageal fistula as acomplication of percutaneous transcatheter ablation of atrial fibrillation.Circulation 2004;109:2724–6.

[79] Martinek M, Meyer C, Hassanein S, et al. Identification of a high-riskpopulation for esophageal injury during radiofrequency catheter ablation ofatrial fibrillation: procedural and anatomical considerations. Heart Rhythm2010;7:1224–30.

[80] Kumagai K, Muraoka S, Mitsutake C, et al. A new approach for completeisolation of the posterior left atrium including pulmonary veins for atrialfibrillation. J Cardiovasc Electrophysiol 2007;18:1047–52.

[81] Aryana A, Heist EK, D’Avila A, et al. Pain and anatomical locations of radio-frequency ablation as predictors of esophageal temperature rise during pulmon-ary vein isolation. J Cardiovasc Electrophysiol 2008;19:32–8.

[82] Chugh A, Rubenstein J, Good E, et al. Mechanical displacement of theesophagus in patients undergoing left atrial ablation of atrial fibrillation.Heart Rhythm 2009;6:319–22.

[83] Kuwahara T, Takahashi A, Kobori A, et al. Safe and effective ablation of atrialfibrillation: importance of esophageal temperature monitoring to avoid

periesophageal nerve injury as a complication of pulmonary vein isolation.J Cardiovasc Electrophysiol 2009;20:1–6.

[84] Di Biase L, Dodig M, Saliba W, et al. Capsule endoscopy in examination ofesophagus for lesions after radiofrequency catheter ablation: a potential toolto select patients with increased risk of complications. J Cardiovasc Electro-

physiol 2010;21:839–44.[85] Yamasaki H, Tada H, Sekiguchi Y, et al. Prevalence and characteristics of

asymptomatic excessive transmural injury after radiofrequency catheterablation of atrial fibrillation. Heart Rhythm 2011;8:826–32.